We are looking for disparate data sources in our company. We have information in several databases that need to be searched on our Intranet. Initial full-text search (FTS) experiments were disappointing. We have implemented a custom search system that works very well for our purposes. Nevertheless, we want to make sure that we are doing the “right thing” and not missing any great tools that would facilitate our work.
What do we need:
- Column Search
- column search capability
- we note which columns in the table are searchable
- Keep some relation between db column and data
- we provide advanced filtering of results
- facilitates (Amazon style) filtering
- provided by grouping the results and allowing the user to filter them using the checkbox
- This is a great feature, users like it.
- Partial Matching Words
- We have many unique identifiers (product identifier, etc.).
- a unique identifier may have subparts with a value (location, etc.).
- or only a part may be available (when the user searches)
- or (by a clearly bad design decision) there may be a space in id
- This is the main function that we have implemented now through CHARINDEX (MSSQL) and INSTR (ORACLE)
- using index functions char turned out to be equivalent performance (+/-) in MSSQL compared to the full text
- not tested Oracle
- however, searching in both types of db is very fast
- We use Indexed (MSSQL) and Materialized (Oracle) views to increase speed.
- . Oracle Materialized , MSSQL.
- (, , , )
- , CTRL-f → →
- ( )
- (. " " )
:
:
- , " " .
- FTS ( )
- - go
- ,
- .
- (+/1) char
- ""
:
, " ", , ?
. , /- nosql (MongoDB .)? (http://www.mongodb.org/display/DOCS/Full+Text+Search+in+Mongo). .
, , - . - " ", . , - , ( , ..) ?
FTS. , , " , , ".